Industries · Canada

AI for Canadian industries

One assistant, shaped around the work each sector actually does — with your data stored in Canada and the models answering on Canadian hardware.

Start chatting Private AI

Why a sector page instead of a generic “AI” page

Most AI marketing says the same thing to everyone. The work does not: a paralegal asks about a matter file, a clinic manager asks about a policy, a plant supervisor asks about a lockout procedure, and each of them needs an answer that cites the document it came from.

So the pages below take the same Canadian foundation and describe what it looks like in one sector at a time — the documents people load first, the questions they ask most, the obligations they have to respect, and the parts of Digby that do the work.

Pick your sector

The six are the sectors we are asked about most. Nothing about Digby is limited to them — a museum, a charity or a surveying firm can use exactly the same projects, knowledge base and scheduled briefings.

Legal — contract review support and matter-file Q&A Healthcare — clinic policy Q&A, not a clinical device Credit unions — member service, policy and regulator framing Public sector — procurement answers and bilingual service Manufacturing — procedures, maintenance and quality standards First Nations — community-controlled data and hosting

The same foundation underneath every sector

Accounts, conversations, memory and uploaded files are stored in Canada (the Canada Central region in Montréal). The models that write answers run on Canadian hardware. Your content is never used to train or fine-tune AI models. And an organisation that needs the whole stack inside its own walls can self-host Digby with its own model server.

What Digby does — and does not — claim Where the models actually run

What every sector setup starts from

01
Projects with a knowledge base
load the policies, manuals, contracts or standards a team works from, and answers come back with the passages they came from.
02
Files you already have
PDFs, Word documents, spreadsheets and images are read directly; a scanned page is read from the page image.
03
Scheduled briefings
a daily or weekly summary of what changed in the documents or inbox a team watches.
04
Connected tools
Gmail, Google Drive and Calendar, Slack, Asana, Xero, GitHub, Discord, Telegram and iMessage, each acting as the person who connected it.
05
A second opinion
contested or high-stakes answers get an independent model check, and “Dig deeper” escalates anything you are about to sign or publish.

See the integrations

Regulated work needs a paper trail

The questions regulators, auditors and procurement teams ask are almost always the same: where is it stored, where is it processed, can we delete it, and can you show us the source of an answer. Digby answers the first three with a location and a delete button, and the fourth with citations on every grounded answer.

What Digby does not have is a certification to hand you. We hold no PIPEDA, Law 25, PHIPA or SOC 2 certificate, and the compliance page says so plainly rather than implying otherwise.

Compliance in plain language The AI-at-work checklist AI for law firms in Toronto AI for law firms in Victoria

Hosted today, self-hosted when you need it

Almost every sector starts hosted: open digby.ai, create a project, upload the documents and invite the team. No deployment project, no procurement cycle.

When the requirement is genuinely “nothing leaves our building”, the same product runs on your own server with Docker Compose and any OpenAI-compatible model endpoint — vLLM, Ollama or LM Studio on your own GPUs.

Self-hosted Digby AI for Canadian businesses

Frequently asked questions

Which industries does Digby work for?

Any organisation that reads documents and writes answers. The six dedicated setups are legal, healthcare, credit unions and financial services, public sector, manufacturing and First Nations administrations — each page describes the documents, questions and obligations of that sector.

Is the AI different per sector?

The underlying assistant is the same. What changes is the project setup — the knowledge base, the instructions a project follows (tone, product names, rules a reply must respect) and the tools connected to it.

Where is our data stored?

In Canada — the Canada Central region in Montréal — covering accounts, conversations, memory and uploaded files.

Do the models run in Canada?

Yes. Digby’s models are served on Canadian-located GPU hardware. A connected app (Gmail, Slack) still talks to that app, and a web search still sends the query to a search service.

Does Digby train on our documents?

No. Your conversations, memory and uploaded files are never used to train or fine-tune AI models.

Can we keep everything on our own infrastructure?

Yes. Self-hosted Digby runs the full stack — database, interface, integrations and coding workspace — on your servers with your own model endpoints.

Is Digby certified for regulated industries?

No. Digby holds no PIPEDA, Law 25, PHIPA or SOC 2 certification, and we do not describe it as compliant. What we can show is where data lives, where inference runs, how deletion works and where each answer came from.